matrixStats.benchmarks

---

colCumprods() and rowCumprods() benchmarks on subsetted computation

This report benchmark the performance of colCumprods() and rowCumprods() on subsetted computation.

Data type “integer”

Data

> rmatrix <- function(nrow, ncol, mode = c("logical", "double", "integer", "index"), range = c(-100, 
+     +100), na_prob = 0) {
+     mode <- match.arg(mode)
+     n <- nrow * ncol
+     if (mode == "logical") {
+         x <- sample(c(FALSE, TRUE), size = n, replace = TRUE)
+     }     else if (mode == "index") {
+         x <- seq_len(n)
+         mode <- "integer"
+     }     else {
+         x <- runif(n, min = range[1], max = range[2])
+     }
+     storage.mode(x) <- mode
+     if (na_prob > 0) 
+         x[sample(n, size = na_prob * n)] <- NA
+     dim(x) <- c(nrow, ncol)
+     x
+ }
> rmatrices <- function(scale = 10, seed = 1, ...) {
+     set.seed(seed)
+     data <- list()
+     data[[1]] <- rmatrix(nrow = scale * 1, ncol = scale * 1, ...)
+     data[[2]] <- rmatrix(nrow = scale * 10, ncol = scale * 10, ...)
+     data[[3]] <- rmatrix(nrow = scale * 100, ncol = scale * 1, ...)
+     data[[4]] <- t(data[[3]])
+     data[[5]] <- rmatrix(nrow = scale * 10, ncol = scale * 100, ...)
+     data[[6]] <- t(data[[5]])
+     names(data) <- sapply(data, FUN = function(x) paste(dim(x), collapse = "x"))
+     data
+ }
> data <- rmatrices(mode = mode, range = c(-1, 1))

Results

10x10 integer matrix

> X <- data[["10x10"]]
> rows <- sample.int(nrow(X), size = nrow(X) * 0.7)
> cols <- sample.int(ncol(X), size = ncol(X) * 0.7)
> X_S <- X[rows, cols]
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5201463 277.8    8529671 455.6  8529671 455.6
Vcells 9793088  74.8   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colCumprods_X_S = colCumprods(X_S), `colCumprods(X, rows, cols)` = colCumprods(X, 
+     rows = rows, cols = cols), `colCumprods(X[rows, cols])` = colCumprods(X[rows, cols]), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5193100 277.4    8529671 455.6  8529671 455.6
Vcells 9765808  74.6   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowCumprods_X_S = rowCumprods(X_S), `rowCumprods(X, cols, rows)` = rowCumprods(X, 
+     rows = cols, cols = rows), `rowCumprods(X[cols, rows])` = rowCumprods(X[cols, rows]), unit = "ms")

Table: Benchmarking of colCumprods_X_S(), colCumprods(X, rows, cols)() and colCumprods(X[rows, cols])() on integer+10x10 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	0.001822	0.0020080	0.0029606	0.0020805	0.0021580	0.088577
2	colCumprods(X, rows, cols)	0.002079	0.0023585	0.0025110	0.0024480	0.0025580	0.004649
3	colCumprods(X[rows, cols])	0.002675	0.0030345	0.0032072	0.0031215	0.0032475	0.007598

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	1.000000	1.000000	1.0000000	1.00000	1.000000	1.0000000
2	colCumprods(X, rows, cols)	1.141054	1.174552	0.8481371	1.17664	1.185357	0.0524854
3	colCumprods(X[rows, cols])	1.468167	1.511205	1.0832998	1.50036	1.504866	0.0857785

Table: Benchmarking of rowCumprods_X_S(), rowCumprods(X, cols, rows)() and rowCumprods(X[cols, rows])() on integer+10x10 data (transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	rowCumprods_X_S	0.001758	0.0020375	0.0021567	0.0021030	0.0021980	0.004234
2	rowCumprods(X, cols, rows)	0.002235	0.0024510	0.0034122	0.0025235	0.0026350	0.087051
3	rowCumprods(X[cols, rows])	0.002617	0.0030700	0.0032103	0.0031590	0.0032695	0.005318

	expr	min	lq	mean	median	uq	max
1	rowCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowCumprods(X, cols, rows)	1.271331	1.202945	1.582154	1.199952	1.198817	20.559991
3	rowCumprods(X[cols, rows])	1.488623	1.506749	1.488543	1.502140	1.487489	1.256023

Figure: Benchmarking of colCumprods_X_S(), colCumprods(X, rows, cols)() and colCumprods(X[rows, cols])() on integer+10x10 data as well as rowCumprods_X_S(), rowCumprods(X, cols, rows)() and rowCumprods(X[cols, rows])() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colCumprods_X_S() and rowCumprods_X_S() on integer+10x10 data (original and transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	1.822	2.0080	2.96063	2.0805	2.158	88.577
2	rowCumprods_X_S	1.758	2.0375	2.15668	2.1030	2.198	4.234

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	1.0000000	1.000000	1.0000000	1.000000	1.000000	1.0000000
2	rowCumprods_X_S	0.9648738	1.014691	0.7284531	1.010815	1.018536	0.0478002

Figure: Benchmarking of colCumprods_X_S() and rowCumprods_X_S() on integer+10x10 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

100x100 integer matrix

> X <- data[["100x100"]]
> rows <- sample.int(nrow(X), size = nrow(X) * 0.7)
> cols <- sample.int(ncol(X), size = ncol(X) * 0.7)
> X_S <- X[rows, cols]
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5191708 277.3    8529671 455.6  8529671 455.6
Vcells 9434675  72.0   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colCumprods_X_S = colCumprods(X_S), `colCumprods(X, rows, cols)` = colCumprods(X, 
+     rows = rows, cols = cols), `colCumprods(X[rows, cols])` = colCumprods(X[rows, cols]), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5191684 277.3    8529671 455.6  8529671 455.6
Vcells 9439728  72.1   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowCumprods_X_S = rowCumprods(X_S), `rowCumprods(X, cols, rows)` = rowCumprods(X, 
+     rows = cols, cols = rows), `rowCumprods(X[cols, rows])` = rowCumprods(X[cols, rows]), unit = "ms")

Table: Benchmarking of colCumprods_X_S(), colCumprods(X, rows, cols)() and colCumprods(X[rows, cols])() on integer+100x100 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
2	colCumprods(X, rows, cols)	0.021441	0.0223445	0.0226823	0.0226510	0.0228050	0.02858
1	colCumprods_X_S	0.022804	0.0238195	0.0241365	0.0239320	0.0240325	0.03798
3	colCumprods(X[rows, cols])	0.033756	0.0351225	0.0356195	0.0352965	0.0354945	0.05809

	expr	min	lq	mean	median	uq	max
2	colCumprods(X, rows, cols)	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colCumprods_X_S	1.063570	1.066012	1.064113	1.056554	1.053826	1.328901
3	colCumprods(X[rows, cols])	1.574367	1.571863	1.570367	1.558276	1.556435	2.032540

Table: Benchmarking of rowCumprods_X_S(), rowCumprods(X, cols, rows)() and rowCumprods(X[cols, rows])() on integer+100x100 data (transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	rowCumprods_X_S	0.023088	0.023602	0.0241635	0.0242950	0.0244940	0.027498
2	rowCumprods(X, cols, rows)	0.029301	0.030197	0.0310790	0.0308155	0.0309935	0.060236
3	rowCumprods(X[cols, rows])	0.034018	0.034710	0.0357621	0.0357075	0.0359090	0.049497

	expr	min	lq	mean	median	uq	max
1	rowCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowCumprods(X, cols, rows)	1.269101	1.279426	1.286195	1.268389	1.265351	2.190559
3	rowCumprods(X[cols, rows])	1.473406	1.470638	1.480001	1.469747	1.466033	1.800022

Figure: Benchmarking of colCumprods_X_S(), colCumprods(X, rows, cols)() and colCumprods(X[rows, cols])() on integer+100x100 data as well as rowCumprods_X_S(), rowCumprods(X, cols, rows)() and rowCumprods(X[cols, rows])() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colCumprods_X_S() and rowCumprods_X_S() on integer+100x100 data (original and transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	22.804	23.8195	24.13649	23.932	24.0325	37.980
2	rowCumprods_X_S	23.088	23.6020	24.16353	24.295	24.4940	27.498

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	1.000000	1.0000000	1.00000	1.000000	1.000000	1.0000000
2	rowCumprods_X_S	1.012454	0.9908688	1.00112	1.015168	1.019203	0.7240126

Figure: Benchmarking of colCumprods_X_S() and rowCumprods_X_S() on integer+100x100 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

1000x10 integer matrix

> X <- data[["1000x10"]]
> rows <- sample.int(nrow(X), size = nrow(X) * 0.7)
> cols <- sample.int(ncol(X), size = ncol(X) * 0.7)
> X_S <- X[rows, cols]
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5192450 277.4    8529671 455.6  8529671 455.6
Vcells 9438730  72.1   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colCumprods_X_S = colCumprods(X_S), `colCumprods(X, rows, cols)` = colCumprods(X, 
+     rows = rows, cols = cols), `colCumprods(X[rows, cols])` = colCumprods(X[rows, cols]), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5192426 277.4    8529671 455.6  8529671 455.6
Vcells 9443783  72.1   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowCumprods_X_S = rowCumprods(X_S), `rowCumprods(X, cols, rows)` = rowCumprods(X, 
+     rows = cols, cols = rows), `rowCumprods(X[cols, rows])` = rowCumprods(X[cols, rows]), unit = "ms")

Table: Benchmarking of colCumprods_X_S(), colCumprods(X, rows, cols)() and colCumprods(X[rows, cols])() on integer+1000x10 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	0.021988	0.0229230	0.0230527	0.0230720	0.023194	0.027323
2	colCumprods(X, rows, cols)	0.022796	0.0238680	0.0243033	0.0242825	0.024512	0.038235
3	colCumprods(X[rows, cols])	0.033491	0.0339615	0.0352297	0.0351105	0.035352	0.060301

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	colCumprods(X, rows, cols)	1.036747	1.041225	1.054248	1.052466	1.056825	1.399371
3	colCumprods(X[rows, cols])	1.523149	1.481547	1.528226	1.521780	1.524187	2.206968

Table: Benchmarking of rowCumprods_X_S(), rowCumprods(X, cols, rows)() and rowCumprods(X[cols, rows])() on integer+1000x10 data (transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	rowCumprods_X_S	0.023211	0.0249755	0.0255337	0.0252190	0.0262985	0.028755
2	rowCumprods(X, cols, rows)	0.031612	0.0339985	0.0348458	0.0352500	0.0357070	0.046902
3	rowCumprods(X[cols, rows])	0.035556	0.0381950	0.0397198	0.0388305	0.0402595	0.070345

	expr	min	lq	mean	median	uq	max
1	rowCumprods_X_S	1.00000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowCumprods(X, cols, rows)	1.36194	1.361274	1.364701	1.397756	1.357758	1.631090
3	rowCumprods(X[cols, rows])	1.53186	1.529299	1.555584	1.539732	1.530867	2.446357

Figure: Benchmarking of colCumprods_X_S(), colCumprods(X, rows, cols)() and colCumprods(X[rows, cols])() on integer+1000x10 data as well as rowCumprods_X_S(), rowCumprods(X, cols, rows)() and rowCumprods(X[cols, rows])() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colCumprods_X_S() and rowCumprods_X_S() on integer+1000x10 data (original and transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	21.988	22.9230	23.05271	23.072	23.1940	27.323
2	rowCumprods_X_S	23.211	24.9755	25.53368	25.219	26.2985	28.755

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.00000
2	rowCumprods_X_S	1.055621	1.089539	1.107622	1.093057	1.133849	1.05241

Figure: Benchmarking of colCumprods_X_S() and rowCumprods_X_S() on integer+1000x10 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

10x1000 integer matrix

> X <- data[["10x1000"]]
> rows <- sample.int(nrow(X), size = nrow(X) * 0.7)
> cols <- sample.int(ncol(X), size = ncol(X) * 0.7)
> X_S <- X[rows, cols]
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5192655 277.4    8529671 455.6  8529671 455.6
Vcells 9439582  72.1   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colCumprods_X_S = colCumprods(X_S), `colCumprods(X, rows, cols)` = colCumprods(X, 
+     rows = rows, cols = cols), `colCumprods(X[rows, cols])` = colCumprods(X[rows, cols]), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5192631 277.4    8529671 455.6  8529671 455.6
Vcells 9444635  72.1   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowCumprods_X_S = rowCumprods(X_S), `rowCumprods(X, cols, rows)` = rowCumprods(X, 
+     rows = cols, cols = rows), `rowCumprods(X[cols, rows])` = rowCumprods(X[cols, rows]), unit = "ms")

Table: Benchmarking of colCumprods_X_S(), colCumprods(X, rows, cols)() and colCumprods(X[rows, cols])() on integer+10x1000 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	0.025506	0.0257285	0.0262958	0.0258245	0.0260210	0.041701
2	colCumprods(X, rows, cols)	0.026499	0.0272175	0.0277237	0.0274540	0.0276560	0.045202
3	colCumprods(X[rows, cols])	0.039103	0.0395035	0.0402056	0.0397990	0.0400215	0.046300

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	colCumprods(X, rows, cols)	1.038932	1.057874	1.054303	1.063099	1.062834	1.083955
3	colCumprods(X[rows, cols])	1.533090	1.535399	1.528976	1.541133	1.538046	1.110285

Table: Benchmarking of rowCumprods_X_S(), rowCumprods(X, cols, rows)() and rowCumprods(X[cols, rows])() on integer+10x1000 data (transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	rowCumprods_X_S	0.021406	0.0218415	0.0221303	0.0219920	0.0222470	0.026927
2	rowCumprods(X, cols, rows)	0.029481	0.0301780	0.0310873	0.0303385	0.0306885	0.057404
3	rowCumprods(X[cols, rows])	0.033306	0.0339000	0.0346626	0.0340470	0.0345015	0.049647

	expr	min	lq	mean	median	uq	max
1	rowCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowCumprods(X, cols, rows)	1.377231	1.381682	1.404741	1.379524	1.379444	2.131838
3	rowCumprods(X[cols, rows])	1.555919	1.552091	1.566297	1.548154	1.550838	1.843763

Figure: Benchmarking of colCumprods_X_S(), colCumprods(X, rows, cols)() and colCumprods(X[rows, cols])() on integer+10x1000 data as well as rowCumprods_X_S(), rowCumprods(X, cols, rows)() and rowCumprods(X[cols, rows])() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colCumprods_X_S() and rowCumprods_X_S() on integer+10x1000 data (original and transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
2	rowCumprods_X_S	21.406	21.8415	22.13029	21.9920	22.247	26.927
1	colCumprods_X_S	25.506	25.7285	26.29576	25.8245	26.021	41.701

	expr	min	lq	mean	median	uq	max
2	rowCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colCumprods_X_S	1.191535	1.177964	1.188225	1.174268	1.169641	1.548669

Figure: Benchmarking of colCumprods_X_S() and rowCumprods_X_S() on integer+10x1000 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

100x1000 integer matrix

> X <- data[["100x1000"]]
> rows <- sample.int(nrow(X), size = nrow(X) * 0.7)
> cols <- sample.int(ncol(X), size = ncol(X) * 0.7)
> X_S <- X[rows, cols]
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5192866 277.4    8529671 455.6  8529671 455.6
Vcells 9462259  72.2   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colCumprods_X_S = colCumprods(X_S), `colCumprods(X, rows, cols)` = colCumprods(X, 
+     rows = rows, cols = cols), `colCumprods(X[rows, cols])` = colCumprods(X[rows, cols]), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5192842 277.4    8529671 455.6  8529671 455.6
Vcells 9512312  72.6   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowCumprods_X_S = rowCumprods(X_S), `rowCumprods(X, cols, rows)` = rowCumprods(X, 
+     rows = cols, cols = rows), `rowCumprods(X[cols, rows])` = rowCumprods(X[cols, rows]), unit = "ms")

Table: Benchmarking of colCumprods_X_S(), colCumprods(X, rows, cols)() and colCumprods(X[rows, cols])() on integer+100x1000 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
2	colCumprods(X, rows, cols)	0.126896	0.1387655	0.1601597	0.1555865	0.1759090	0.277073
1	colCumprods_X_S	0.134134	0.1492715	0.1720807	0.1678155	0.1934280	0.246906
3	colCumprods(X[rows, cols])	0.200267	0.2291715	0.2606636	0.2541380	0.2841425	0.356816

	expr	min	lq	mean	median	uq	max
2	colCumprods(X, rows, cols)	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
1	colCumprods_X_S	1.057039	1.075711	1.074432	1.078599	1.099591	0.8911226
3	colCumprods(X[rows, cols])	1.578198	1.651502	1.627524	1.633419	1.615281	1.2878050

Table: Benchmarking of rowCumprods_X_S(), rowCumprods(X, cols, rows)() and rowCumprods(X[cols, rows])() on integer+100x1000 data (transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	rowCumprods_X_S	0.128142	0.149196	0.1677644	0.1627060	0.181955	0.244547
2	rowCumprods(X, cols, rows)	0.167780	0.182036	0.2084298	0.2041705	0.219153	0.327482
3	rowCumprods(X[cols, rows])	0.191850	0.221605	0.2487560	0.2418305	0.277303	0.325854

	expr	min	lq	mean	median	uq	max
1	rowCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowCumprods(X, cols, rows)	1.309329	1.220113	1.242395	1.254843	1.204435	1.339137
3	rowCumprods(X[cols, rows])	1.497167	1.485328	1.482770	1.486303	1.524020	1.332480

Figure: Benchmarking of colCumprods_X_S(), colCumprods(X, rows, cols)() and colCumprods(X[rows, cols])() on integer+100x1000 data as well as rowCumprods_X_S(), rowCumprods(X, cols, rows)() and rowCumprods(X[cols, rows])() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colCumprods_X_S() and rowCumprods_X_S() on integer+100x1000 data (original and transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
2	rowCumprods_X_S	128.142	149.1960	167.7644	162.7060	181.955	244.547
1	colCumprods_X_S	134.134	149.2715	172.0807	167.8155	193.428	246.906

	expr	min	lq	mean	median	uq	max
2	rowCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colCumprods_X_S	1.046761	1.000506	1.025728	1.031403	1.063054	1.009646

Figure: Benchmarking of colCumprods_X_S() and rowCumprods_X_S() on integer+100x1000 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

1000x100 integer matrix

> X <- data[["1000x100"]]
> rows <- sample.int(nrow(X), size = nrow(X) * 0.7)
> cols <- sample.int(ncol(X), size = ncol(X) * 0.7)
> X_S <- X[rows, cols]
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5193076 277.4    8529671 455.6  8529671 455.6
Vcells 9463043  72.2   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colCumprods_X_S = colCumprods(X_S), `colCumprods(X, rows, cols)` = colCumprods(X, 
+     rows = rows, cols = cols), `colCumprods(X[rows, cols])` = colCumprods(X[rows, cols]), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5193052 277.4    8529671 455.6  8529671 455.6
Vcells 9513096  72.6   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowCumprods_X_S = rowCumprods(X_S), `rowCumprods(X, cols, rows)` = rowCumprods(X, 
+     rows = cols, cols = rows), `rowCumprods(X[cols, rows])` = rowCumprods(X[cols, rows]), unit = "ms")

Table: Benchmarking of colCumprods_X_S(), colCumprods(X, rows, cols)() and colCumprods(X[rows, cols])() on integer+1000x100 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
2	colCumprods(X, rows, cols)	0.121293	0.1392520	0.1606841	0.1574875	0.1692505	0.237311
1	colCumprods_X_S	0.128511	0.1474575	0.1661782	0.1616625	0.1711695	0.251502
3	colCumprods(X[rows, cols])	0.193243	0.2196735	0.2546033	0.2427595	0.2880905	0.374580

	expr	min	lq	mean	median	uq	max
2	colCumprods(X, rows, cols)	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
1	colCumprods_X_S	1.059509	1.058925	1.034192	1.026510	1.011338	1.059799
3	colCumprods(X[rows, cols])	1.593192	1.577525	1.584495	1.541452	1.702155	1.578435

Table: Benchmarking of rowCumprods_X_S(), rowCumprods(X, cols, rows)() and rowCumprods(X[cols, rows])() on integer+1000x100 data (transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	rowCumprods_X_S	0.134418	0.1417770	0.1634373	0.158983	0.1761790	0.232506
2	rowCumprods(X, cols, rows)	0.170420	0.1795645	0.2088120	0.207295	0.2220780	0.350441
3	rowCumprods(X[cols, rows])	0.201386	0.2090830	0.2436314	0.231745	0.2669365	0.343087

	expr	min	lq	mean	median	uq	max
1	rowCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowCumprods(X, cols, rows)	1.267836	1.266528	1.277628	1.303881	1.260525	1.507234
3	rowCumprods(X[cols, rows])	1.498207	1.474731	1.490673	1.457672	1.515144	1.475605

Figure: Benchmarking of colCumprods_X_S(), colCumprods(X, rows, cols)() and colCumprods(X[rows, cols])() on integer+1000x100 data as well as rowCumprods_X_S(), rowCumprods(X, cols, rows)() and rowCumprods(X[cols, rows])() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colCumprods_X_S() and rowCumprods_X_S() on integer+1000x100 data (original and transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
2	rowCumprods_X_S	134.418	141.7770	163.4373	158.9830	176.1790	232.506
1	colCumprods_X_S	128.511	147.4575	166.1782	161.6625	171.1695	251.502

	expr	min	lq	mean	median	uq	max
2	rowCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.0000000	1.000000
1	colCumprods_X_S	0.956055	1.040066	1.016771	1.016854	0.9715659	1.081701

Figure: Benchmarking of colCumprods_X_S() and rowCumprods_X_S() on integer+1000x100 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

Data type “double”

Data

> rmatrix <- function(nrow, ncol, mode = c("logical", "double", "integer", "index"), range = c(-100, 
+     +100), na_prob = 0) {
+     mode <- match.arg(mode)
+     n <- nrow * ncol
+     if (mode == "logical") {
+         x <- sample(c(FALSE, TRUE), size = n, replace = TRUE)
+     }     else if (mode == "index") {
+         x <- seq_len(n)
+         mode <- "integer"
+     }     else {
+         x <- runif(n, min = range[1], max = range[2])
+     }
+     storage.mode(x) <- mode
+     if (na_prob > 0) 
+         x[sample(n, size = na_prob * n)] <- NA
+     dim(x) <- c(nrow, ncol)
+     x
+ }
> rmatrices <- function(scale = 10, seed = 1, ...) {
+     set.seed(seed)
+     data <- list()
+     data[[1]] <- rmatrix(nrow = scale * 1, ncol = scale * 1, ...)
+     data[[2]] <- rmatrix(nrow = scale * 10, ncol = scale * 10, ...)
+     data[[3]] <- rmatrix(nrow = scale * 100, ncol = scale * 1, ...)
+     data[[4]] <- t(data[[3]])
+     data[[5]] <- rmatrix(nrow = scale * 10, ncol = scale * 100, ...)
+     data[[6]] <- t(data[[5]])
+     names(data) <- sapply(data, FUN = function(x) paste(dim(x), collapse = "x"))
+     data
+ }
> data <- rmatrices(mode = mode, range = c(-1, 1))

Results

10x10 double matrix

> X <- data[["10x10"]]
> rows <- sample.int(nrow(X), size = nrow(X) * 0.7)
> cols <- sample.int(ncol(X), size = ncol(X) * 0.7)
> X_S <- X[rows, cols]
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5193294 277.4    8529671 455.6  8529671 455.6
Vcells 9554158  72.9   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colCumprods_X_S = colCumprods(X_S), `colCumprods(X, rows, cols)` = colCumprods(X, 
+     rows = rows, cols = cols), `colCumprods(X[rows, cols])` = colCumprods(X[rows, cols]), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5193261 277.4    8529671 455.6  8529671 455.6
Vcells 9554296  72.9   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowCumprods_X_S = rowCumprods(X_S), `rowCumprods(X, cols, rows)` = rowCumprods(X, 
+     rows = cols, cols = rows), `rowCumprods(X[cols, rows])` = rowCumprods(X[cols, rows]), unit = "ms")

Table: Benchmarking of colCumprods_X_S(), colCumprods(X, rows, cols)() and colCumprods(X[rows, cols])() on double+10x10 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	0.001723	0.0018920	0.0021445	0.0019985	0.002053	0.016670
2	colCumprods(X, rows, cols)	0.001984	0.0022735	0.0023828	0.0023190	0.002411	0.004820
3	colCumprods(X[rows, cols])	0.002756	0.0029670	0.0031282	0.0030700	0.003185	0.007519

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
2	colCumprods(X, rows, cols)	1.151480	1.201639	1.111126	1.160370	1.174379	0.2891422
3	colCumprods(X[rows, cols])	1.599536	1.568182	1.458694	1.536152	1.551388	0.4510498

Table: Benchmarking of rowCumprods_X_S(), rowCumprods(X, cols, rows)() and rowCumprods(X[cols, rows])() on double+10x10 data (transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	rowCumprods_X_S	0.001694	0.0019545	0.0022622	0.0020285	0.0021635	0.012304
2	rowCumprods(X, cols, rows)	0.002157	0.0023070	0.0027454	0.0023850	0.0025010	0.033352
3	rowCumprods(X[cols, rows])	0.002732	0.0030520	0.0032582	0.0031625	0.0033055	0.008278

	expr	min	lq	mean	median	uq	max
1	rowCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
2	rowCumprods(X, cols, rows)	1.273318	1.180353	1.213584	1.175746	1.155997	2.7106632
3	rowCumprods(X[cols, rows])	1.612751	1.561525	1.440225	1.559034	1.527848	0.6727893

Figure: Benchmarking of colCumprods_X_S(), colCumprods(X, rows, cols)() and colCumprods(X[rows, cols])() on double+10x10 data as well as rowCumprods_X_S(), rowCumprods(X, cols, rows)() and rowCumprods(X[cols, rows])() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colCumprods_X_S() and rowCumprods_X_S() on double+10x10 data (original and transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	1.723	1.8920	2.14450	1.9985	2.0530	16.670
2	rowCumprods_X_S	1.694	1.9545	2.26225	2.0285	2.1635	12.304

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	1.0000000	1.000000	1.000000	1.000000	1.000000	1.0000000
2	rowCumprods_X_S	0.9831689	1.033034	1.054908	1.015011	1.053824	0.7380924

Figure: Benchmarking of colCumprods_X_S() and rowCumprods_X_S() on double+10x10 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

100x100 double matrix

> X <- data[["100x100"]]
> rows <- sample.int(nrow(X), size = nrow(X) * 0.7)
> cols <- sample.int(ncol(X), size = ncol(X) * 0.7)
> X_S <- X[rows, cols]
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5193493 277.4    8529671 455.6  8529671 455.6
Vcells 9560122  73.0   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colCumprods_X_S = colCumprods(X_S), `colCumprods(X, rows, cols)` = colCumprods(X, 
+     rows = rows, cols = cols), `colCumprods(X[rows, cols])` = colCumprods(X[rows, cols]), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5193469 277.4    8529671 455.6  8529671 455.6
Vcells 9570175  73.1   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowCumprods_X_S = rowCumprods(X_S), `rowCumprods(X, cols, rows)` = rowCumprods(X, 
+     rows = cols, cols = rows), `rowCumprods(X[cols, rows])` = rowCumprods(X[cols, rows]), unit = "ms")

Table: Benchmarking of colCumprods_X_S(), colCumprods(X, rows, cols)() and colCumprods(X[rows, cols])() on double+100x100 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	0.012336	0.0128960	0.0132486	0.0131285	0.0133315	0.022399
2	colCumprods(X, rows, cols)	0.012199	0.0130155	0.0139613	0.0134930	0.0138705	0.044148
3	colCumprods(X[rows, cols])	0.028018	0.0291685	0.0299914	0.0294915	0.0296575	0.054085

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	1.0000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	colCumprods(X, rows, cols)	0.9888943	1.009266	1.053798	1.027764	1.040431	1.970981
3	colCumprods(X[rows, cols])	2.2712387	2.261825	2.263747	2.246372	2.224618	2.414617

Table: Benchmarking of rowCumprods_X_S(), rowCumprods(X, cols, rows)() and rowCumprods(X[cols, rows])() on double+100x100 data (transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	rowCumprods_X_S	0.016491	0.0171385	0.0176957	0.017725	0.0180225	0.023410
2	rowCumprods(X, cols, rows)	0.017039	0.0177355	0.0186849	0.018173	0.0187380	0.043777
3	rowCumprods(X[cols, rows])	0.031488	0.0329055	0.0340712	0.033576	0.0344720	0.049186

	expr	min	lq	mean	median	uq	max
1	rowCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.00000	1.000000
2	rowCumprods(X, cols, rows)	1.033230	1.034834	1.055899	1.025275	1.03970	1.870013
3	rowCumprods(X[cols, rows])	1.909405	1.919976	1.925396	1.894274	1.91272	2.101068

Figure: Benchmarking of colCumprods_X_S(), colCumprods(X, rows, cols)() and colCumprods(X[rows, cols])() on double+100x100 data as well as rowCumprods_X_S(), rowCumprods(X, cols, rows)() and rowCumprods(X[cols, rows])() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colCumprods_X_S() and rowCumprods_X_S() on double+100x100 data (original and transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	12.336	12.8960	13.24858	13.1285	13.3315	22.399
2	rowCumprods_X_S	16.491	17.1385	17.69571	17.7250	18.0225	23.410

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowCumprods_X_S	1.336819	1.328978	1.335668	1.350116	1.351873	1.045136

Figure: Benchmarking of colCumprods_X_S() and rowCumprods_X_S() on double+100x100 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

1000x10 double matrix

> X <- data[["1000x10"]]
> rows <- sample.int(nrow(X), size = nrow(X) * 0.7)
> cols <- sample.int(ncol(X), size = ncol(X) * 0.7)
> X_S <- X[rows, cols]
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5193692 277.4    8529671 455.6  8529671 455.6
Vcells 9561538  73.0   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colCumprods_X_S = colCumprods(X_S), `colCumprods(X, rows, cols)` = colCumprods(X, 
+     rows = rows, cols = cols), `colCumprods(X[rows, cols])` = colCumprods(X[rows, cols]), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5193668 277.4    8529671 455.6  8529671 455.6
Vcells 9571591  73.1   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowCumprods_X_S = rowCumprods(X_S), `rowCumprods(X, cols, rows)` = rowCumprods(X, 
+     rows = cols, cols = rows), `rowCumprods(X[cols, rows])` = rowCumprods(X[cols, rows]), unit = "ms")

Table: Benchmarking of colCumprods_X_S(), colCumprods(X, rows, cols)() and colCumprods(X[rows, cols])() on double+1000x10 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	0.013771	0.0144340	0.0155603	0.0145785	0.0149540	0.028134
2	colCumprods(X, rows, cols)	0.015254	0.0161860	0.0175489	0.0165900	0.0168765	0.033195
3	colCumprods(X[rows, cols])	0.030178	0.0313385	0.0326032	0.0315635	0.0318470	0.062813

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	colCumprods(X, rows, cols)	1.107690	1.121380	1.127797	1.137977	1.128561	1.179889
3	colCumprods(X[rows, cols])	2.191417	2.171158	2.095283	2.165072	2.129664	2.232637

Table: Benchmarking of rowCumprods_X_S(), rowCumprods(X, cols, rows)() and rowCumprods(X[cols, rows])() on double+1000x10 data (transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	rowCumprods_X_S	0.018863	0.0197505	0.0208456	0.0203745	0.0211705	0.035212
2	rowCumprods(X, cols, rows)	0.022238	0.0233740	0.0246005	0.0243440	0.0252385	0.040047
3	rowCumprods(X[cols, rows])	0.036841	0.0380115	0.0406848	0.0395765	0.0412700	0.081365

	expr	min	lq	mean	median	uq	max
1	rowCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowCumprods(X, cols, rows)	1.178922	1.183464	1.180129	1.194827	1.192154	1.137311
3	rowCumprods(X[cols, rows])	1.953083	1.924584	1.951724	1.942453	1.949411	2.310718

Figure: Benchmarking of colCumprods_X_S(), colCumprods(X, rows, cols)() and colCumprods(X[rows, cols])() on double+1000x10 data as well as rowCumprods_X_S(), rowCumprods(X, cols, rows)() and rowCumprods(X[cols, rows])() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colCumprods_X_S() and rowCumprods_X_S() on double+1000x10 data (original and transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	13.771	14.4340	15.56029	14.5785	14.9540	28.134
2	rowCumprods_X_S	18.863	19.7505	20.84560	20.3745	21.1705	35.212

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowCumprods_X_S	1.369762	1.368332	1.339667	1.397572	1.415708	1.251582

Figure: Benchmarking of colCumprods_X_S() and rowCumprods_X_S() on double+1000x10 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

10x1000 double matrix

> X <- data[["10x1000"]]
> rows <- sample.int(nrow(X), size = nrow(X) * 0.7)
> cols <- sample.int(ncol(X), size = ncol(X) * 0.7)
> X_S <- X[rows, cols]
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5193897 277.4    8529671 455.6  8529671 455.6
Vcells 9561674  73.0   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colCumprods_X_S = colCumprods(X_S), `colCumprods(X, rows, cols)` = colCumprods(X, 
+     rows = rows, cols = cols), `colCumprods(X[rows, cols])` = colCumprods(X[rows, cols]), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5193873 277.4    8529671 455.6  8529671 455.6
Vcells 9571727  73.1   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowCumprods_X_S = rowCumprods(X_S), `rowCumprods(X, cols, rows)` = rowCumprods(X, 
+     rows = cols, cols = rows), `rowCumprods(X[cols, rows])` = rowCumprods(X[cols, rows]), unit = "ms")

Table: Benchmarking of colCumprods_X_S(), colCumprods(X, rows, cols)() and colCumprods(X[rows, cols])() on double+10x1000 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	0.014407	0.0149005	0.0154546	0.0150870	0.0152950	0.040956
2	colCumprods(X, rows, cols)	0.015947	0.0173390	0.0186091	0.0177910	0.0181915	0.050309
3	colCumprods(X[rows, cols])	0.033177	0.0347320	0.0358207	0.0350095	0.0354080	0.058166

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	colCumprods(X, rows, cols)	1.106893	1.163652	1.204115	1.179227	1.189376	1.228367
3	colCumprods(X[rows, cols])	2.302839	2.330929	2.317806	2.320508	2.315005	1.420207

Table: Benchmarking of rowCumprods_X_S(), rowCumprods(X, cols, rows)() and rowCumprods(X[cols, rows])() on double+10x1000 data (transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	rowCumprods_X_S	0.015177	0.0158065	0.0161303	0.0160155	0.0163330	0.020969
2	rowCumprods(X, cols, rows)	0.017840	0.0186015	0.0195433	0.0189065	0.0191330	0.049796
3	rowCumprods(X[cols, rows])	0.030701	0.0319855	0.0327561	0.0323630	0.0334105	0.046895

	expr	min	lq	mean	median	uq	max
1	rowCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowCumprods(X, cols, rows)	1.175463	1.176826	1.211587	1.180513	1.171432	2.374744
3	rowCumprods(X[cols, rows])	2.022864	2.023566	2.030715	2.020730	2.045583	2.236397

Figure: Benchmarking of colCumprods_X_S(), colCumprods(X, rows, cols)() and colCumprods(X[rows, cols])() on double+10x1000 data as well as rowCumprods_X_S(), rowCumprods(X, cols, rows)() and rowCumprods(X[cols, rows])() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colCumprods_X_S() and rowCumprods_X_S() on double+10x1000 data (original and transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	14.407	14.9005	15.45457	15.0870	15.295	40.956
2	rowCumprods_X_S	15.177	15.8065	16.13031	16.0155	16.333	20.969

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.0000000
2	rowCumprods_X_S	1.053446	1.060803	1.043724	1.061543	1.067865	0.5119885

Figure: Benchmarking of colCumprods_X_S() and rowCumprods_X_S() on double+10x1000 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

100x1000 double matrix

> X <- data[["100x1000"]]
> rows <- sample.int(nrow(X), size = nrow(X) * 0.7)
> cols <- sample.int(ncol(X), size = ncol(X) * 0.7)
> X_S <- X[rows, cols]
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5194108 277.4    8529671 455.6  8529671 455.6
Vcells 9607144  73.3   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colCumprods_X_S = colCumprods(X_S), `colCumprods(X, rows, cols)` = colCumprods(X, 
+     rows = rows, cols = cols), `colCumprods(X[rows, cols])` = colCumprods(X[rows, cols]), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5194084 277.4    8529671 455.6  8529671 455.6
Vcells 9707197  74.1   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowCumprods_X_S = rowCumprods(X_S), `rowCumprods(X, cols, rows)` = rowCumprods(X, 
+     rows = cols, cols = rows), `rowCumprods(X[cols, rows])` = rowCumprods(X[cols, rows]), unit = "ms")

Table: Benchmarking of colCumprods_X_S(), colCumprods(X, rows, cols)() and colCumprods(X[rows, cols])() on double+100x1000 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	0.077247	0.0830300	0.0924106	0.0881235	0.0994515	0.138448
2	colCumprods(X, rows, cols)	0.080875	0.0880115	0.0962571	0.0921245	0.1035620	0.217449
3	colCumprods(X[rows, cols])	0.188139	0.2061475	0.2239854	0.2153790	0.2455685	0.298529

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	colCumprods(X, rows, cols)	1.046966	1.059996	1.041624	1.045402	1.041332	1.570619
3	colCumprods(X[rows, cols])	2.435551	2.482807	2.423805	2.444059	2.469229	2.156254

Table: Benchmarking of rowCumprods_X_S(), rowCumprods(X, cols, rows)() and rowCumprods(X[cols, rows])() on double+100x1000 data (transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
2	rowCumprods(X, cols, rows)	0.099090	0.1048165	0.1187806	0.1147640	0.1244035	0.250598
1	rowCumprods_X_S	0.096603	0.1103330	0.1206186	0.1180620	0.1296205	0.178054
3	rowCumprods(X[cols, rows])	0.200336	0.2136050	0.2409966	0.2345775	0.2572995	0.364135

	expr	min	lq	mean	median	uq	max
2	rowCumprods(X, cols, rows)	1.0000000	1.000000	1.000000	1.000000	1.000000	1.0000000
1	rowCumprods_X_S	0.9749016	1.052630	1.015473	1.028737	1.041936	0.7105164
3	rowCumprods(X[cols, rows])	2.0217580	2.037895	2.028922	2.043999	2.068266	1.4530643

Figure: Benchmarking of colCumprods_X_S(), colCumprods(X, rows, cols)() and colCumprods(X[rows, cols])() on double+100x1000 data as well as rowCumprods_X_S(), rowCumprods(X, cols, rows)() and rowCumprods(X[cols, rows])() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colCumprods_X_S() and rowCumprods_X_S() on double+100x1000 data (original and transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	77.247	83.030	92.41065	88.1235	99.4515	138.448
2	rowCumprods_X_S	96.603	110.333	120.61855	118.0620	129.6205	178.054

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowCumprods_X_S	1.250573	1.328833	1.305245	1.339733	1.303354	1.286071

Figure: Benchmarking of colCumprods_X_S() and rowCumprods_X_S() on double+100x1000 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

1000x100 double matrix

> X <- data[["1000x100"]]
> rows <- sample.int(nrow(X), size = nrow(X) * 0.7)
> cols <- sample.int(ncol(X), size = ncol(X) * 0.7)
> X_S <- X[rows, cols]
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5194318 277.5    8529671 455.6  8529671 455.6
Vcells 9607285  73.3   31876688 243.2 60562128 462.1
> colStats <- microbenchmark(colCumprods_X_S = colCumprods(X_S), `colCumprods(X, rows, cols)` = colCumprods(X, 
+     rows = rows, cols = cols), `colCumprods(X[rows, cols])` = colCumprods(X[rows, cols]), unit = "ms")
> X <- t(X)
> X_S <- t(X_S)
> gc()
          used  (Mb) gc trigger  (Mb) max used  (Mb)
Ncells 5194294 277.5    8529671 455.6  8529671 455.6
Vcells 9707338  74.1   31876688 243.2 60562128 462.1
> rowStats <- microbenchmark(rowCumprods_X_S = rowCumprods(X_S), `rowCumprods(X, cols, rows)` = rowCumprods(X, 
+     rows = cols, cols = rows), `rowCumprods(X[cols, rows])` = rowCumprods(X[cols, rows]), unit = "ms")

Table: Benchmarking of colCumprods_X_S(), colCumprods(X, rows, cols)() and colCumprods(X[rows, cols])() on double+1000x100 data. The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	0.085267	0.0931515	0.1010496	0.0970570	0.1044525	0.153772
2	colCumprods(X, rows, cols)	0.085064	0.0916330	0.1016335	0.0982665	0.1060800	0.196777
3	colCumprods(X[rows, cols])	0.191793	0.2081295	0.2299603	0.2231495	0.2477090	0.388096

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	1.0000000	1.0000000	1.000000	1.000000	1.000000	1.000000
2	colCumprods(X, rows, cols)	0.9976192	0.9836986	1.005779	1.012462	1.015581	1.279667
3	colCumprods(X[rows, cols])	2.2493227	2.2343118	2.275717	2.299159	2.371499	2.523840

Table: Benchmarking of rowCumprods_X_S(), rowCumprods(X, cols, rows)() and rowCumprods(X[cols, rows])() on double+1000x100 data (transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	rowCumprods_X_S	0.108001	0.1172790	0.1363158	0.1340380	0.1458500	0.187411
2	rowCumprods(X, cols, rows)	0.106519	0.1212385	0.1372090	0.1343805	0.1440325	0.289116
3	rowCumprods(X[cols, rows])	0.203520	0.2234670	0.2569952	0.2501560	0.2743865	0.356161

	expr	min	lq	mean	median	uq	max
1	rowCumprods_X_S	1.0000000	1.000000	1.000000	1.000000	1.0000000	1.000000
2	rowCumprods(X, cols, rows)	0.9862779	1.033761	1.006552	1.002555	0.9875386	1.542684
3	rowCumprods(X[cols, rows])	1.8844270	1.905431	1.885293	1.866307	1.8812924	1.900427

Figure: Benchmarking of colCumprods_X_S(), colCumprods(X, rows, cols)() and colCumprods(X[rows, cols])() on double+1000x100 data as well as rowCumprods_X_S(), rowCumprods(X, cols, rows)() and rowCumprods(X[cols, rows])() on the same data transposed. Outliers are displayed as crosses. Times are in milliseconds.

Table: Benchmarking of colCumprods_X_S() and rowCumprods_X_S() on double+1000x100 data (original and transposed). The top panel shows times in milliseconds and the bottom panel shows relative times.

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	85.267	93.1515	101.0496	97.057	104.4525	153.772
2	rowCumprods_X_S	108.001	117.2790	136.3158	134.038	145.8500	187.411

	expr	min	lq	mean	median	uq	max
1	colCumprods_X_S	1.000000	1.000000	1.000000	1.000000	1.000000	1.000000
2	rowCumprods_X_S	1.266621	1.259014	1.348999	1.381023	1.396329	1.218759

Figure: Benchmarking of colCumprods_X_S() and rowCumprods_X_S() on double+1000x100 data (original and transposed). Outliers are displayed as crosses. Times are in milliseconds.

Appendix

Session information

R version 4.1.1 Patched (2021-08-10 r80727)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Ubuntu 18.04.5 LTS

Matrix products: default
BLAS:   /home/hb/software/R-devel/R-4-1-branch/lib/R/lib/libRblas.so
LAPACK: /home/hb/software/R-devel/R-4-1-branch/lib/R/lib/libRlapack.so

locale:
 [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C              
 [3] LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8    
 [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8   
 [7] LC_PAPER=en_US.UTF-8       LC_NAME=C                 
 [9] LC_ADDRESS=C               LC_TELEPHONE=C            
[11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C       

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
[1] microbenchmark_1.4-7   matrixStats_0.60.1     ggplot2_3.3.5         
[4] knitr_1.33             R.devices_2.17.0       R.utils_2.10.1        
[7] R.oo_1.24.0            R.methodsS3_1.8.1-9001 history_0.0.1-9000    

loaded via a namespace (and not attached):
 [1] Biobase_2.52.0          httr_1.4.2              splines_4.1.1          
 [4] bit64_4.0.5             network_1.17.1          assertthat_0.2.1       
 [7] highr_0.9               stats4_4.1.1            blob_1.2.2             
[10] GenomeInfoDbData_1.2.6  robustbase_0.93-8       pillar_1.6.2           
[13] RSQLite_2.2.8           lattice_0.20-44         glue_1.4.2             
[16] digest_0.6.27           XVector_0.32.0          colorspace_2.0-2       
[19] Matrix_1.3-4            XML_3.99-0.7            pkgconfig_2.0.3        
[22] zlibbioc_1.38.0         genefilter_1.74.0       purrr_0.3.4            
[25] ergm_4.1.2              xtable_1.8-4            scales_1.1.1           
[28] tibble_3.1.4            annotate_1.70.0         KEGGREST_1.32.0        
[31] farver_2.1.0            generics_0.1.0          IRanges_2.26.0         
[34] ellipsis_0.3.2          cachem_1.0.6            withr_2.4.2            
[37] BiocGenerics_0.38.0     mime_0.11               survival_3.2-13        
[40] magrittr_2.0.1          crayon_1.4.1            statnet.common_4.5.0   
[43] memoise_2.0.0           laeken_0.5.1            fansi_0.5.0            
[46] R.cache_0.15.0          MASS_7.3-54             R.rsp_0.44.0           
[49] progressr_0.8.0         tools_4.1.1             lifecycle_1.0.0        
[52] S4Vectors_0.30.0        trust_0.1-8             munsell_0.5.0          
[55] tabby_0.0.1-9001        AnnotationDbi_1.54.1    Biostrings_2.60.2      
[58] compiler_4.1.1          GenomeInfoDb_1.28.1     rlang_0.4.11           
[61] grid_4.1.1              RCurl_1.98-1.4          cwhmisc_6.6            
[64] rappdirs_0.3.3          startup_0.15.0          labeling_0.4.2         
[67] bitops_1.0-7            base64enc_0.1-3         boot_1.3-28            
[70] gtable_0.3.0            DBI_1.1.1               markdown_1.1           
[73] R6_2.5.1                lpSolveAPI_5.5.2.0-17.7 rle_0.9.2              
[76] dplyr_1.0.7             fastmap_1.1.0           bit_4.0.4              
[79] utf8_1.2.2              parallel_4.1.1          Rcpp_1.0.7             
[82] vctrs_0.3.8             png_0.1-7               DEoptimR_1.0-9         
[85] tidyselect_1.1.1        xfun_0.25               coda_0.19-4            

Total processing time was 22.71 secs.

Reproducibility

To reproduce this report, do:

html <- matrixStats:::benchmark('colRowCumprods_subset')

---

Copyright Dongcan Jiang. Last updated on 2021-08-25 18:53:31 (+0200 UTC). Powered by RSP.